Adopting a number of perspectives, this paper returns to explore the cornerstone in Saussure’s work of the arbitrariness of the sign. The paper systematically considers the sign’s arbitrariness in the following aspects.
https://doi.org.10.33234/JSAR.2.3
pp. 50-82
Returning to Saussure: The Sign is Arbitrary in all Possible Manners
Thomas Olver
unaffiliated scholar
Pretoria, South Africa
olver@gmx.ch
Abstract
Adopting a number of perspectives, this paper returns to explore the cornerstone in Saussure’s work of the arbitrariness of the sign. The paper systematically considers the sign’s arbitrariness in the following aspects: language as an endowment of human evolution; language in its two dimensions, langue and parole; the absence of any referent for the sign; the sign’s twofold structure of signifier and signified; the bond between signifier and signified; the floating realms of sound and thought out of which this structure emerges; the evolutionary precedents to these two realms; and the role played by convention in the formation and persistence of the sign. Absolute arbitrariness in all these aspects is demonstrated, and confirms that the sign is arbitrary in all possible manners.
Keywords: Saussure, language, arbitrariness, sign, signifier, signified, convention, evolution
The bond between the signifier and the signified is arbitrary. Since I mean by sign the whole that results from the associating of the signifier with the signified, I can simply say: the linguistic sign is arbitrary. (Saussure, Course, p. 67)
Introduction
Many commentators and scholars report difficulty in understanding Saussure’s ideas. David Holdcroft (1991, p. 1), for example, opens a lengthy study of Saussure with the following observation: “The Course in General Linguistics is not an easy book to read.” In some studies, difficulties and confusion arise around a specific idea, such as the role or field of convention (e.g. Gvoždiak 2022). The arbitrariness of the sign appears to present special difficulty for any number of scholars. Only a few decades after the publication of the Course, no less a semiotician than Émile Benveniste makes the polemic claim that “entre le signifiant et le signifié, le lien n’est pas arbitraire; au contraire, il est nécessaire” [“the bond between the signifier and the signified is not arbitrary; on the contrary it is necessary“] (1939, p. 25; own translation; emphasis in original). Certainly, the bond is necessary, but certainly not “on the contrary”; the bond between the signifier and the signified becomes necessary only after it has been established, arbitrarily. It is remarkable that Benveniste’s bald-faced sleight of hand should have become so influential, although it must be said that the apotheosis of repudiating the arbitrariness of the sign is surely Bolinger (1949) under the influence of Behaviourism in the USA, transposed to linguistics by the influential work of Bloomfield (1933) in particular. To this day, a host of scholars continues to deny arbitrariness (e.g. Gamkrelidze 1974; Simone 1995; Nanny and Fischer 1999; Perniss et al. 2010; Thompson et al. 2012; Dingemanse et al. 2015; Abramova & Fernandez 2016; Stoltz 2021; Danesi 2022), and the originality of Saussure’s ideas is often reduced to a straw man against which such scholarship seeks to establish its anti-arbitrariness credentials.
The trend of repudiating arbitrariness has been accelerated by the emergence of a fully-fledged iconicity theory of language, with its roots in the Socratic dialogue Cratylus (Plato, c. 384-3) and the opposition between convention and naturalism (spoken for by Hermogenes and Cratylus, respectively). In modernity, the iconicity theory of language has been underpinned by the Pragmatic philosophy of Charles Sanders Peirce (e.g. i. 1893-1902, i. 1897-1910; see also the robust synthesis of Peirce’s thinking by Jakobson 1965; by Honders & Honders 1982; and by Cuypere 2008) so that iconicity has now generally come to mean non-arbitrariness (Downing & Stiebels 2012). Curiously, however, Jakobson (1965, p. 37, quoting Peirce: “Every word is a symbol. Every sentence is a symbol. Every book is a symbol…”) appears to conclude that Peirce himself repudiates iconicity!
The discovery of Saussure’s so-called Orangery Manuscripts in 1996 (now included in the Writings) presents an additional source of difficulty for scholars and has triggered substantial debate (e.g. Stawarska 2015, 2020) around differences and (in)consistencies between these authored yet unpublished manuscripts and the un-authored Course (including other editions of student Course notes), with the suggestion even being put forward that the ideas in these manuscripts of Saussure have a remarkable resemblance to the phenomenology of Peirce. While in my view (as will be seen below), there is no material conceptual divergence between the Writings and the Course, especially on the topic of the arbitrariness of the sign, it is not my purpose here to demonstrate this in any detail. Furthermore, and following the approach of Roland Barthes (1967) and Michel Foucault (1969), I make no claim to oracular insight into Saussure’s supposed authorial intentions in either the Writings or the lectures that provided the material for the Course, nor do I believe that fidelity to the traditionally conceived singular author is a relevant criterion to judge the ideas encountered in the Writings and the Course (or indeed any set of abstract ideas, such as the copious – unedited and unpublished – manuscripts of Peirce). Asserting such authorial fidelity has value only to copyright lawyers and authoritarian leaders. The value of an idea is not determined by its genesis but by its ability to explore and explain reality.
In a manner parallel to the theory of iconicity, some scholars (e.g. Jakobson 1957, 1965; Silverstein 1976) make an argument for an indexical theory of language, with the focus, for example, on the logic of syntax, tense and time, although this seems to broaden significantly Peirce’s notion of an index, perhaps facilitated by the blurring of index and icon in diagrammatic iconicity (i.e. sequence, e.g. Ceaser’s famous veni, vidi, vici). Even overlooking such conceptual hybridity, the role of time in language is highly complex, and must be viewed against the baseline of a primeval human state of timelessness, which knows no temporal or sequential logic. In this earliest state, one of the constitutive functions of language is its introduction of time, which in both ontogenesis and phylogenesis does not strictly exist for humans before the advent of language. Here it is only necessary to consider Freud’s principle of the timelessness of the unconscious (1915, p. 187) where the id, the initial and most primitive component of the human mental apparatus, does not know time: “There is nothing in the id that corresponds to the idea of time; there is no recognition of the passage of time, and – a thing that is most remarkable and awaits consideration in philosophical thought – no alteration in its mental processes is produced by the passage of time” (1933, p. 74). Only later in the child’s life is time introduced, and not in the id but in the ego, via the advent of language and the narrative function made possible by language (Author 2025, pp. 137-140). There is no reason to think the same sequence (timelessness before the advent of language) is not true in human phylogenesis or, in principle, that an equivalent constitutive function of language should not apply for spatial logic.
It must be noted that iconicity and indexicality theories of language and signs are generally – and perhaps by necessity – blind to the obvious social and conventional nature of icons and indexes. This blind spot is demonstrated, for example, in a rich study of iconicity in theories of morphology over a great number of unrelated languages (Downing and Stiebels 2012) that never once pauses to consider the role of convention in the manifestations of iconicity it presents. In my view, any quality of iconicity or indexicality in signs is derived from their conventionality, which in turn is always a co-function of arbitrariness. In this view, the possibility or presence of iconicity and indexicality is acknowledged, but they are only ever recognisable a posteriori to convention, they are not necessary qualities of the sign a priori. No doubt this view will be hard to accept when seen from within the Peircean paradigm of perfect signs and their balanced “coassistance of the three functions” (Jakobson 1965, p. 26), but – in the intellectual spirit of opening up fresh perspectives – I trust my discussion below will afford a refreshing and stimulating perspective.
Unsurprisingly, indexicality is modest in its theoretical claims. On closer examination, the more voluminous theory of iconicity turns out to rely on the conceit of very limited and specific examples, such as the beloved icon stock of onomatopoeia and so-called motivated signs, but both these categories are easily dispatched as conventions in the Course. Physical gesture and its more formalised system of standard sign language is another favoured exemplar of iconicity theory, even though gestures and signing are by their nature culturally specific and explicitly conventional. Furthermore, they are not so much signs or a language as an encoding of natural language in a medium different to sound, comparable in its coding effect to Morse code or even – in principle – the written form of language. Such coding systems are therefore doubly conventional, a fact overlooked by many scholars, perhaps because they are exclusively or predominantly concerned with the communicative function of language (e.g. Jakobson 1960, 1985; Silverstein 1976, Lakoff & Johnson 1980; Sebeok 1994), which in turn often rests on a somewhat naïve theory of language as a simple mirror of the world, with language as the coding or modelling of a solid reality, the “object” or referent of Peirce’s tripartite sign. This is to neglect the fundamentally constitutive character of language. The constitutive function of language in the absence of any object for humans that is independent of language underpins the following discussion.
The sign is arbitrary in all possible manners
In both the Writings and the Course, Saussure’s ideas are sublime in their simplicity and reasonableness. Even the open-ended and provisional nature of Saussure’s formulations – a function also in part of their genesis in medias res during his lectures and in the notes taken of these lectures by several students, as much as Saussure’s own incomplete documented record – invites the reader to enter into and contribute to the evolving understanding of his ideas. The following presentation seeks to emulate this simplicity and straightforwardness in a return to Saussure. If there is an inherent difficulty in understanding and talking about Saussure’s ideas, and especially the concept of the arbitrariness of the sign, it is that we must understand and talk about the sign using the sign. Perhaps it is this difficulty that scholars mean when they highlight the complexity of Saussure’s ideas. Perhaps this necessary reflexivity may also explain – at least in part – the resistance of Saussure’s ideas to being formulated in his own words (the Writings) or in the words of others (the Course) – a resistance that still plays a role today in the reception of Saussure’s ideas.
The reflexivity of using signs to understand and talk about those very signs is a clue to the first step in the process of returning to Saussure. Later on in the discussion, I will call it the step of nomination, but the Course and the Writings use other words: viewpoint and starting point:
Far from it being the object that antedates the viewpoint, it would seem that it is the viewpoint that creates the object; besides, nothing tells us in advance that one way of considering the fact in question takes precedence over the others or is in any way superior to them. (Course, p. 8)
There are five or six basic truths which are so interrelated that it is equally possible to use any one as the starting point, and to arrive logically at all the others and at every minute ramification of the consequences, starting from any one of them. (Writings, p. 3)
Let us remember in fact that the object in linguistics does not exist to start with, is not predetermined in its own right. Hence to speak of an object, to name an object, is nothing more than to invoke a particular point of view. (Writings, p. 8)
This multiplicity of viewpoints evokes Nietzsche’s concept of Perspectivism in epistemology:
Soweit überhaupt das Wort “Erkenntniß” Sinn hat, ist die Welt erkennbar: aber sie ist anders deutbar, sie hat keinen Sinn hinter sich, sondern unzählige Sinne “Perspektivismus”.
Unsre Bedürfnisse sind es, die die Welt auslegen: unsre Triebe und deren Für und Wider. Jeder Trieb ist eine Art Herrschsucht, jeder hat seine Perspektive, welche er als Norm allen übrigen Trieben aufzwingen möchte. (1885-1887, p. 315)
[Insofar as the word “knowledge” is appropriate to use in this context, the world is knowable. The world can be interpreted in multiple ways; it has no meaning per se, but an infinite number of meanings. This is “Perspectivism”.
It is our needs that determine how we interpret the world: our drives and their conflicts. Every drive is a form of hunger for power; each has its perspective that seeks to prevail over all other drives, establishing itself as the norm. (own translation)]
The thing to understand is: there will always be a specific perspective, and not only as a starting point. The crucial thing is to embrace this understanding and to acknowledge from the start the perspective that is chosen as the starting point, as well as later perspectives that are added. Here is the first grip on the arbitrariness of the sign. There will always be a privileged perspective, selected from the many – perhaps even infinite – perspectives from which to look at the structure of the sign. The act of selecting a perspective is an act of nomination, where even the many – perhaps even infinite – perspectives as much as the act of selection are clearly also a manifestation and operative element of the arbitrariness at work here. For a variety of reasons, this choice, this particular perspective, may be selected unconsciously, or a particular perspective may become so orthodox, so naturalised, so conventionalised by its repeated selection, that the very act of selecting, of nominating, the perspective becomes a so-called fact: so self-evident and so obvious it is blind and blinding to all other perspectives. Here, in the act of nomination, lies the fundamental randomness of the sign, but the nominee is often oblivious to the randomness – the arbitrariness – and thus one more difficulty in understanding and thinking about the arbitrariness of the sign arises. To recognise and see the arbitrariness of the sign means also to recognise and see the arbitrariness of perspective and nomination, the inescapable incompleteness of perspective, and the arbitrariness of convention itself.
One investigative strategy is to escape perspectives. Not an escape from perspective per se, because that is impossible. There will always be a perspective. No, the strategy is to escape from the naturalised perspective, as a start, and to develop the approach of acknowledging and interrogating the current perspective. Then, aim to escape from mono-nomination, from a singular perspective, and to develop the strategy of poly-perspectives. They are there, they are waiting to be tried out. A strategy of shifting perspectives, of trying on new perspectives, allows for an escape from the illusory belief that there is only one perspective. A synonym for perspective is ideology, where ideology can be compared to glass: transparent and reflective – both equally useful analogies. All of this is to say that the same problem – the question of perspective – can be approached in different ways, in order to de-familiarise (Shklovsky 1917) the commonplace, the seemingly natural, the obvious perspective.
Nomination is paired with arbitrariness. Confronted by the arbitrariness of the sign, it is valid and necessary to start from – to nominate – a particular perspective, but it is essential to understand that this nomination is doomed to fail if it becomes fixed, if one perspective out of many is nominated and other perspectives are stifled or neglected. There is no escape in mono-nomination. Only a plurality of perspectives will allow for an understanding of the arbitrariness of the sign, because the sign is arbitrary in all possible manners, and there is no inherently fixed aspect of the sign, so a fixed perspective will inevitably blind the viewer to arbitrariness, by making the sign seem fixed, in even one aspect. If we are to understand and talk about the arbitrariness of the sign – and, what is more, to do this using signs – then the first step, the opening perspective, has to be the nomination of perspectives in the plural.
Here immediately arises the first objection to this approach: if we admit plurality at the start – plurality of perspectives – then we admit arbitrariness – the very thing we wish to demonstrate.
A self-serving argument, if ever there was one. The only reply, the deflection of mono-nomination, is to reiterate the necessity for acknowledging and interrogating perspective if we are ever to escape to the relative freedom of multi- and poly-perspectives. There are those who wish to hold onto their mono-perspectives, and so be it, but they sacrifice all the other perspectives, and they are left poorer for their sacrifice. They will also struggle to make any advance in understanding and talking about the arbitrariness of the sign; they will find it difficult, even futile, and they should trouble themselves no further.
In adopting the starting position, the nominated perspective, that there are a multitude of perspectives, we are now well positioned to look at the sign from a number of perspectives, and yet every perspective will lead us back to and confirm the arbitrariness of the sign, and we will see that the sign is arbitrary in all possible manners. This will be the case as soon as a second perspective is added, and each additional perspective will make it clearer and easier to understand and talk about the arbitrariness of the sign. To reach the core of the discussion of the arbitrariness of the sign, it is necessary to start first with language, as the exemplary category of the sign. By necessity, this will embrace language in its two dimensions: langue and parole. From this double dimension, the essence of the sign’s two parts – signifier and signified – and the absence of any referent can then emerge in the discussion of arbitrariness. To reach this destination, a starting point further back in the perspective will help: the characteristic of language as an endowment of human evolution. At this earlier starting point – this nomination – language is joined by the two other great gifts of human evolution, namely labour and sexual reproduction. Here too arbitrariness rules supreme. In the conclusion, it will even be possible to form a synthesis of perspectives, giving the composite view that a fly might have with its compound eyes.
The Endowments of Evolution
Taking the starting perspective of the three primary endowments of human evolution allows us to see language –and the linguistic sign – inclusively in the broader human perspective, along with labour and sexual reproduction. There is a basic and fundamental materiality to language that is easily overlooked. In this context of evolution, language is material in the same way as labour and libido. This perspective recognises that language, labour and libido is vested in the body, and allows us to understand and talk about this trinity as part of the arbitrary action which has made human evolution possible. In each of these three aspects, the evolution of the human body has preordained an armature in the endowment that has a double movement. Evolution proceeds by the simple logic of function, and yet humans combine a second factor with function, namely an excess to that which is strictly necessary for the evolutionary function. Both armatures, function and excess, are characterised by arbitrariness.
Arbitrariness is a feature of evolution itself, shown in the supreme arbitrariness of selection combined with the logic of function, where function is directly equal to survival – of the species, and therefore of specific traits. Here the synchronic view and diachronic view show different perspectives; in the synchronic view, arbitrariness is obvious, whereas the diachronic view allows the apparent importance of the function to be conspicuous, with an ostensibly inherent reason – cause and effect – that seems to stand in contrast to arbitrariness. The diachronic importance of the function is only apparent, however, and for two reasons. First, because the apparent or nominal function of a trait is always subordinate to the overriding function of survival of the species. Survival – of the species as well as traits – is the ultimate and only logic at work, and yet survival is ultimately arbitrary in the extreme, always contingent on a multitude of variable factors; the variation of even one of these factors by a small degree can lead to immediate failure of survival of the species, showing how function is not any less arbitrary than the factors giving rise to function. An obvious example is the arbitrariness of sexual versus asexual reproduction; in fact, there is inevitably a combination of both types in many species, if the asexual reproduction of cells is also considered, showing that each type of reproduction has a functional benefit, although this function cannot hide the utter arbitrariness of the type of reproduction, and the arbitrary introduction of sexual reproduction at some point in the evolution of life. Of course, there is then also the utter arbitrariness to consider which is demonstrated in each individual act of gamete fertilisation, where the pairing of two gametes is as random as is possible. Each coming together of gametes in the act of fertilisation recapitulates the arbitrariness of evolution itself. The outcome of fertilisation is also random, limited only by the system of combination possibilities of the underlying genetic material.
In the case of human labour, work in the broadest sense is a necessary function for survival, where even in the most benign tropical setting the food that humans need may be at hand but still has to be collected and prepared for consumption. This early hunter-gatherer labour is necessary but arbitrary in its focus, determined in situ by the range of environmental settings where the human species has spread. Particular constellations of arbitrary labour are merely entrenched by permanent settlement in a location. Labour is further characterised by arbitrariness in the division of labour that humans introduce, thereby producing the categories of class. In the evolutionary view, only the labour of conceiving and bearing children is biologically limited in its arbitrariness to the female of the species by the evolutionary selection of sexual reproduction, and yet humans have gone on to develop complex rules and practices for the division of labour, so that labour becomes the most significant factor in human social structure, after biological sex, despite the utter arbitrariness of both.
A second reason for the function of these three primary endowments being only superficially important is shown in a modulated and bifurcated perspective. In very broad terms, sexual reproduction, labour and language are the three pillars that characterise and distinguish human evolution. Each of these traits or factors clearly plays a decisive functional role in human evolution, and yet the original evolutionary functions are almost unrecognisable in modern humans, where any number of non-functional and even dysfunctional attachments to the three original functional traits – individually or in their interplay as the formation of human society – now modulate the original function so that a bifurcated perspective of two armatures is revealed. The bifurcated perspective makes it possible to recognise that humans introduce a novel element to the three fundamental functions bequeathed by evolution. This new element is the pursuit of a surplus, an excess. Surplus is understood here in the purely relative sense of activity (activation of the evolutionary trait) that is in excess of what would be necessary for the simple performance of the functions in question. That the excess is non-functional or even dysfunctional in the evolutionary perspective makes it possible to characterise these surplus pursuits as fetishes. Freud (1905, 1927) defines a fetish as an attachment that is a substitute for an original object or function, but he goes on to show how the defining character of a fetish is not only the substitution but also the concomitant disavowal of the original object or function, even as it is replaced by the substitute. Disavowal is the necessary prerequisite for a substitute to become necessary and emerge. Together disavowal and substitution – both unconscious forces – form the second armature: the common fetishistic factor in humans that is the substitution and subsequent attachment to a surplus which humans produce in all three evolutionary areas. The surplus has its roots in the functional activation of the original trait but it is surplus – redundant – to the function itself. Attachment to the surplus as substitute becomes the manifest and defining social form of the three primary evolutionary traits. While the pursuit of pleasure may seem to be the motive underlying the new fetish, pleasure explains only the substitution and not the preceding disavowal, which must be directed at some aspect of the original evolutionary function.
Examples of the fetish surplus can be considered in all three traits to uncover the motive for disavowal. The trait of sexual reproduction in humans is replete with examples of surplus, so much so that Freud (1905) introduces the fundamental distinction between procreative and non-procreative sexual activity. Based on this distinction, Freud’s entire body of work sets out to show in various ways how the non-procreative sphere of sexuality is core to the development not only of an individual’s character but also of human society in general, especially in the form of the repression of sexuality which is a universal characteristic of human societies. An excess of sexuality to procreation, very often in the repressed form of the libido’s unconscious action in the mind, is the defining feature of modern humans. However, there is another more direct example of excess to consider in the field of procreation, and that is the unconstrained exponential quantitative growth of the human population, to the degree that it now threatens the sustainability of humans and of all other life on the planet.
In the field of labour, the fetish excess is easy to witness – consider only the nuclear bomb or space travel to the moon. These have no valid function for survival; on the contrary, they are futile and deleterious in the extreme to survival. Objectively speaking, they are contraindicated for survival, and yet they played a central role in the historical development of human society in the 20th century, albeit as a continuation of earlier expressions of war and expansion, and they continue to threaten human survival (nuclear war) and consume vast resources in unproductive displays of excess (space travel). War in general is the epitome of useless dysfunctional excess labour. In a far more detailed manner, Marx (1867) has crystallised and explored the idea of excess labour in the form of alienated accumulated labour as the driving force of capital and capitalism, also with a view to understanding the ways in which the unmitigated excess of slavery has been replaced in the modern economy via the intricacies of wage labour as a means of extracting surplus value. From a more hopeful perspective, Marx has characterised this bifurcation as a realm of necessity and a realm of freedom, where freedom in a very practical sense results from a shortening of the working day to the time required to meet physical necessities, thereby leaving time in the day for “that development of human energy which is an end in itself” (1863-1883, p. 807). In this perspective, Marx is echoing the “purposiveness without a purpose” of Kant’s aesthetic theory (1790, p. 57), where Kant too relies on a bifurcated distinction between objective purposiveness (necessity, utility) and subjective purposiveness (excess, beauty) (p. 45).
Language is the most recent and so perhaps least primal of the three endowments of evolution, and yet it is also replete with excess. While the function of communication was an important factor driving the emergence of language in evolution, it would be a fallacy and conceit to reduce human language to this single function. At the very least, the mnemonic function of language is just as important in its contribution to the evolution and survival of homo sapiens. Not only does the mnemonic function of language directly produce an excess in the material development of the large human brain, it produces the conditions to stimulate further the expansion of neural and cognitive activity. This is precisely an example of the surplus that is produced by language, literally in an organic, material sense. Brain and memory co-evolve, along with the voice mechanism, to become capable of more and more elaborate sounds with scope for ever-greater sound differentiation and ever-greater memory function, in a reinforcing succession. The invention of writing is the predictable next stage of excess in language. While predictable, the invention of writing is hardly inevitable, as is demonstrated by the many cultures that never developed their own system of writing and relied instead on a purely oral tradition. At some tipping point of social and economic development, however, writing becomes essential because it makes possible a further and far greater expansion of the range of the mnemonic function, especially in the increasing complexity of social and economic relationships and their organisation, graphically manifesting and demonstrating the constitutive function of language. Writing also enhances the possibility of a standard version of a language emerging by that version becoming the written form of record, the form necessary to learn to read and write.
The perspective of a double armature downgrades the claim of the primacy of the communication function of language by the traditional schools of linguistics and philosophy of language, and places the function of communication in comparative relation to other evolutionary functions of language that have often been neglected. Even in the communicative function, the bifurcated perspective reveals how language ends up exceeding the evolutionary origin: it produces a surplus that may have its roots in communication but which goes on to function in an intra-subject as well as inter-subject form of communication. While the intra-subjectivity of language is still arguably a form of communication, this is not what is typically meant when scholars evoke language as communication, and yet the communicative excess that is intra-subjectivity is a crucial aspect in the development of consciousness in homo sapiens. This excess is the human subject’s signification of itself to itself. It could perhaps be called a primary phatic expression made possible by language, and in which is found the original foundation of consciousness – at least in the sense of being conscious of the self.
Language functions in other ways to produce a surplus and a redundancy of communication so that the communication function – already always imprecise, provisional and approximate – becomes detached or is removed from language in the way it is deployed by humans. Growing from the mnemonic function of language, an obvious example of this surplus is the aesthetic use of language: patterns and associations of sounds (rhyme, intonation, emphasis, accents) and thoughts that produce redundancy of sound (poetry) and meaning (narrative). It is also interesting to see in poetry the affinity created by the interplay of paradigmatic associations, while narrative is the result of an accumulation of syntagmatic associations. Both kinds of association – paradigmatic and syntagmatic – can play a role in metaphor. Such affinities are the effect of clustering that produces the aesthetic character of language, although these aesthetic features are tightly related to the mnemonic function. While perhaps not an original function of language in human evolution per se, the importance of the aesthetic element in language as part of the broader function of memory has a pre-eminent place as well as an appeal of its own in the originality of parole it makes possible. Easier to identify as a surplus, rhetoric and propaganda are language functions with a clear excess – even redundancy – for communication, and they demonstrate an absence of truth function. While some sort of elementary truth function would presumably have been kernel to the evolutionary origin of language as communication, in order to establish mutual trust, the ubiquitous human use of rhetoric and propaganda in various forms and media shows how little language or communication has to do with truth.
These excess functions in the three primary evolutionary traits open the door to a further consideration, namely the perspective that not only language but also sexual reproduction and labour can be viewed as unconscious social systems (langues) that make possible a number of functions, and these functions are all examples of paroles, individual expressions only made possible within the system, in exactly the way that the distinction for language between langue and parole in the Course shows (pp. 13-14). Here the arbitrariness of the sign will also be carried over from the langue to the paroles.
The conclusion for all three traits is that the primary evolutionary function (sexual – procreation; labour – survival; language – communication and memory) is eclipsed by the human fetishistic attachment to a surplus obtained. In each case, the straight evolutionary function is augmented by some surplus dimension of pleasure and accumulation, or pleasure through accumulation. It is interesting to ask why this unconsumed accumulation and surplus should happen in the first place, though there is no easy answer. Plato’s Symposium suggests the human drive is characterised by a basic pursuit of objects not already possessed, and in turn the extension of this drive in the form of accumulation is thus the desire to remain in possession of the object in the future: “I wish I may continue to have in the future what I already have at present” (p. 35). Although not stated as such in Plato’s text, this basic possessive and accumulative drive can be understood as the underpinning of the concept of surplus in the extra-evolutionary sense of the double armature. It also serves as a concise definition of power: the power to possess and accumulate, which in turn can be seen as the expression of a fundamental human need for the illusion of immortality.
Herein lies the underlying motive for disavowal: there is in human action a fundamental disavowal of their own mortality. Paradoxically, the path of evolution for humans – indeed for any species – has led to an indirect form of immortality in the survival of the species, but for every individual of the species, the path leads always to death. For even the most long-lived species, time runs out for the individual. As Marx (1847, p. 30) puts it so eloquently: “Le temps est tout, l’homme n’est plus rien; il est tout au plus la carcasse du temps” [“Time is everything, humans are nothing; they are, at most, time’s carcase”, own translation]. In place of the primary evolutionary traits, which lead directly to individual mortality, humans disavow mortality and substitute the armature of surplus as the fetish and illusion of immortality. By either illusion – the direct illusion of immortality or the indirect illusion of power as a route of achieving immortality – the armature of surplus is a fundamental disavowal of the finiteness of human time.
In this way, the human species is unique, so that the disavowal of time and its substitution with illusions may be said to be the special human preserve. In this way too, the arbitrariness of the sign makes itself known. Of all life, only humans struggle against the primary logic of evolution. Only humans pursue the illusion of immortality through procreation; only humans seek to amass the surplus labour of others in the vain belief it will stave off their own death, even as this stolen labour becomes their mausoleum; and only humans have the illusion of mastery over time that is activated and maintained in language, especially the mnemonic function of language. Time is unmoved though, and is resolute in its contempt for these arbitrary illusions. Language is frequently characterised in the Course and the Writings as a chain, and we can say – to echo Marx – that language is time’s chain, chaining humans to time.
The diachronic perspective on language is much favoured in scholarship because it allows the scholar to take hold of something seemingly concrete and seemingly not arbitrary: the origins of a particular word or language or aspect of language. Predictably, this quest to identify the diachronic lineage of language echoes the lineage that humans accord themselves in family trees: Z was born of parents X and Y, who in turn were born of their parents, and so on, reaching back to the point where the difficulty of accurate record keeping makes a mockery of the family origin myth. While X and Y are the biological parents of Z, this does not release any of them in any way from the utter arbitrariness of sexual reproduction in the first place, and in all dimensions; it merely lends a feeling of legitimacy and stasis to the arbitrariness of the originary sign. Even as the history of individuals and of languages becomes increasingly speculative and mythical the further back the origins are traced, the genealogical and etymological excavations allow a certain kind of authority to be stamped on a particular nomination, seemingly relieving it of the original sin of arbitrariness. In turn, though, the conjured authority of the past is embraced and propounded as a mantle of present authority and certainty by those who pursue this kind of linguistic or biological archaeology. They rely on diachronic authority to avoid confronting the utter arbitrariness of the sign. No doubt it is interesting to see the origin of words or languages or ideas or peoples in other words, languages, ideas, peoples, but the seemingly precise historical cleavages that this diachronic scholarship imputes do not in reality exist, neither in languages nor indeed in the peoples of the earth or their ideas. As the example (Writings, pp. 100-102) of the relationship between Latin and French illustrates, “the notion of separate languages becomes a mere notion of [geographical and chronological] distance” with a “chain linking the great units of Indo-European” (p. 223).
There is only continuity in the development over time of signs and languages and ideas and peoples in an ever-receding lineage. The evolution of species offers the perfect instructive analogy: while the science of biology today produces great taxonomies of species and their development – one from another in a vast family tree of life – the plain situation is that life itself does not know or recognise these taxonomies and distinctions of lineage. There is only one physical distinction of any consequence and importance: life and non-life. All of life, as differentiated and multiform as life may be, is entirely homogenous when seen in the distinction to non-life. How fragile life is, how easily it is extinguished, and how fiercely it struggles to survive. The situation of language and signs is the same. There are signs – distinct, different, apparently a multifold – and yet they pale in comparison to the silence that is the non-sign. Language is as fragile as life itself, and as arbitrary.
Signifier and signified
Advancing then to consider the sign in detail, the full range of the arbitrariness of the sign culminates in the components: signifier and signified. The sign is fundamentally arbitrary in the following ways: in its twofold structure of signifier and signified; in the bond between signifier and signified; in the floating realms of sound and thought out of which this structure emerges; in the precedents to these two realms; and in the role played by convention in the formation and persistence of the sign.
The starting point for a definition of the sign is the twofold structure described in the Course: the sign is the signifier and the signified (p. 67). To start with the most basic understanding of this sign structure as the bond between a sound (the signifier) and a mental image (the signified), the first aspect of arbitrariness relates to the bond between the two. This bond – the joining together – is entirely arbitrary. Even instances of so-called motivated signs, such as those discussed in the Course (pp. 131-134), are only motivated in the very narrow sense of being derivatives of other linguistic signs (i.e. relative motivation), and any claim to their motivation rests entirely on this derivation: “motivation is never absolute” (p. 132), with a so-called motivated sign always derived from another sign or signs. In no way does this derivation negate or undo the arbitrariness of the initial sign(s), and the original arbitrariness is simply transferred and incorporated into the derivation. If anything, the derivative relationship reinforces the fact of arbitrariness by highlighting the role that convention plays in the arbitrariness of the sign, because convention is the direct coefficient of arbitrariness in the linguistic system, and in any other system of signs.
An effective definition of convention is: the social adoption and repetition of a signifier-signified bond such that it can become a sign. Convention does the work of making the sign by soldering the signifier to the signified so that they become a sign. There can be no sign without the social dimension of convention. Naturalising the signifier-signified bond is central to the action of convention. Even though the sign is always entirely arbitrary, it takes on the guise of being self evident through its conventional use. This presents something that may appear like a paradox: the arbitrary signifier-signified bond becomes apparently fixed. It is this paradox which foxes someone like Benveniste, where the success of the work of convention can make it so difficult to recognise and accept the arbitrariness of the sign. Undercutting a position like Benveniste’s, the Course expresses the situation succinctly and lyrically: “Because the sign is arbitrary, it follows no law other than that of tradition, and because it is based on tradition, it is arbitrary” (p. 74). Convention (“tradition”) is simply the glue that holds the signifier and the signified together – it is the bond, formed by the accretion of usage over time that lends the sign a feeling of stability and solidity. Yet it is not the sign that is stable or solid, merely the convention cement – however fragile it may turn out to be – that holds the signifier to the signified, and the signified to the signifier, for it is a bi-directional mutual bond in arbitrariness.
Confronted in a sober perspective, there is no paradox, although it may feel like a paradox in the moment when the arbitrariness of the sign is first confronted. Convention as the social dimension of the sign is the very dimension in which users are socially immersed and which may blind them to convention. Signs are by definition social, i.e. conventional – the two words are synonymous. Outside of the social dimension, there is no sign. Convention takes the two elements of the sign and binds them in a social vise, turning an otherwise incomprehensible sound and an otherwise indistinct thought into the slice of sound and thought that are paired to each other in a pattern which becomes totally familiar and recognisable to the users of the sign system.
More can be said about convention and the work of convention in tandem with arbitrariness. Two important components of convention are differentiation and repetition, which will be discussed presently. First, however, seen from the reflexive perspective of a discussion about the sign using signs, it is possible to say that convention is also nomination, albeit unconscious to the speaker. This word – nomination – helps to foreground and defamiliarise the action of convention, which freezes a particular arbitrariness through social usage – repetition – and thereby manufactures the sign. This is how the apparent paradox is resolved. Although nomination suggests a conscious action on the part of the users of the convention, it is not at all a process of conscious selection. If there is any conscious action involved in the use of language, it is re-selection: as they speak (parole), speakers continually re-select from the pre-established system (langue) of previous liaisons between signifier and signified, the origin of which is lost in the mists of time. That origin too could not have been conscious, since in principle it predates consciousness in the modern sense. To the extent that a language user has any choice in the matter, the choice – nomination – by a speaker is not the choice to bring a signifier and signified together at the start, at the birth of the sign, but the choice to keep using a sign (in parole) once it has been born. The birth – the origin of a sign or a language – is lost in the past; and while the origin of signs and languages is of enormous consequence for the very concept of the past and time, the birth of a sign or a language is of no consequence to its current users, the vast majority of whom are severely ignorant of the history of their own language. To language users, all that matters is that once upon a time this signifier and this signified became joined as a sign. All that matters now is that the signifier-signified union is maintained, however mutably this may occur in practice (pp. 71-78, Course); and the signifier-signified union is maintained through usage, passed on from one generation to the next; and it is this usage that is convention: the ongoing, repeated nomination of an arbitrary bond, repetition which lends the sign sufficient permanence and an apparent sensibility that is necessarily beyond the control of individual speakers, who are simply born into the language community as infants or later join a language community through the far more arduous work of learning a second language.
In summary so far, the bond between signifier and signified is arbitrary, and in both directions. Arbitrariness is overlaid with the character of a convention which is always a nomination. This draws attention to a further aspect of arbitrariness, namely the absence of referentiality in the sign. Reference, or the referent as it is sometimes called in the case of individual signs, is a topic of much confusion. There is often a childlike need to link the sign with a referent, and to see this link as somehow essential and inherent. This infantile urge to impute referents is probably due to a mnemonic effect that is the result of methods often employed in the early stages of language learning to provide an initial linguistic structure (Dick and Jane; the cat sat on the mat, etc.). Perhaps because these early associations are laid down in the early years of a child’s life, they take on such power later, and lend strength – by this force of early association and attachment (i.e. identification) – to arguments for the referentiality or non-arbitrariness of the sign. The same unconscious identification applies to attachments to old aesthetic patterns and forms found in poetry and narrative in particular, and even the cultural privileging granted the written form of language. The argument for referentiality thus relies on a persistent unconscious attachment to relics of the infant period, and is one of the most frequent and common errors encountered in linguistics and the philosophy of language. This referential or realistic perspective relies on unconscious childhood memories of the artificial scaffolding provided in language learning, comparable to the overpowering and lasting strength of the child’s earliest relations to its parents and other caregivers and authority figures that endure in the structural affinities these relations establish in the unconscious mental apparatus.
As a general observation, it must be remembered: the only link that exists is the bond between signifier and signified. There are two further aspects to consider here. First, the attribution of a sign to a referent is entirely imputed post factum – there is no actual link, only perhaps the impression and expectation of a link bestowed by the arbitrary but naturalised (i.e. conventionalised) bond between signifier and signified. This bond is naturalised in the social world of convention. The referent – to the extent that it is even possible to impute such an entity – is simply a function of the arbitrary pair bond of the sign in its social environment. In certain literary contexts, this has been called vraisemblance or verisimilitude: the ability to conjure up an apparently realistic scene; but this meaning can be extended as a function to all signs due to their fundamentally social nature. Given the arbitrariness of the sign, there is not, nor could there ever be, any link to a referent, let alone one that is inherent or innate. On the contrary, the category of referent is a by-product of the arbitrariness and conventionality of the sign: the sign constitutes the world. This is the second aspect to consider here. Once the reference or referent is recognised as merely a function of the convention, the category of referent can be understood as a mirage – a false representation of empirical reality. This is not to say that there could ever be a true representation of empirical reality, however. There can only ever be at best a cloudy contact zone between the senses on one hand and empirical reality on the other. This is only partly related to the famous image of the floating realms of sound and thought in the Course (p. 113), which is another way of illustrating the arbitrariness of the sign. The floating realms will be discussed in a moment, especially in the way these realms come together to form signs; first, however, there is an anteroom to the floating realms, and an overarching arbitrariness that precedes the firmer yet still entirely arbitrary ground of sound and thought bonded in signs.
In principle, all living organisms are always self-enclosed, and therefore cut off from their broader physical (empirical) environment, in which they nevertheless exist. As a result, each species of organism must therefore develop sensory mechanisms with which to navigate intercourse with the broader environment. These mechanisms include sensory openings to allow for sampling of the environment in order to determine the suitability of material for nourishment, for example, or to identify risk and danger. It is through these senses that the organism is able to establish and maintain contact with the surrounding physical environment – empirical reality – but the senses already represent a first degree of intermediation between the organism and empirical reality via the physically delimited perspective given by the senses. Much is lost or altered in the intermediation, even as perception of certain aspects of the external world may also be enhanced – such as danger – based on patterns, the perception of which through the senses is gradually laid down in the process of evolution as a condition of survival. In turn, these patterns form the basis for memory, which in humans serves as a neural matrix upon which ideas take root, grow and become the floating realm of thought. Perception – this anteroom of thought – is also characterised by extreme arbitrariness, which is demonstrated by the multifarious evolutionary development of species and their varied sensory mechanisms, from basic pseudopodia in the smallest protozoa to the increasingly more sophisticated sensory cells and organs of plants and animals. This variation in development can be explained by the variation in the local conditions in which life evolved, but in principle both the development and the environment are entirely arbitrary. Whether life was born once or many times is immaterial; it multiplied and evolved over eons, spreading in an ebb and flow to new environments, new empirical realities, which in turn defined the parameters for survival and shaped the subsequent adaptations. In this way, life itself in the synchronic view is analogous to langue, with a multitude of evolutionary paroles; and in the diachronic view, life also changes in a way that is analogous to language change: the ancient forms are documented in preserved records, like fossils, but the present forms also retain and show evidence and remnants of their preceding forms in their synchronic manifestations.
The point here is to acknowledge the principle of arbitrariness not only in the origin of the species but also in the origins of the human organs of sound and thought, as well as ultimately the signs that are the product of the bond between sound and thought. The human brain, the seat of thought, and its pairing with the human vocal apparatus is an entirely arbitrary outcome of evolution. Similar structures in analogous combinations but with different outcomes can be seen to some degree in other species, especially those closest to humans in the evolutionary taxonomies but, by the same token, in the process of evolution humans have also lost sensitivity in certain sensory endowments that serve a purpose similar to language in other species, such as the sense of smell and taste, notably in combination with pheromone production. The particular human constellation of producing sounds and thoughts in combination is, to begin with, an entirely arbitrary outcome of evolution, and there is nothing fixed or predetermined in the constellations of language or in the human capacity for language as an evolutionary outcome. In a similar manner, the host organs of language – the brain in particular – can be enhanced or stunted in each individual in their particular development, quite apart from any physiological impediment, simply by use or neglect. The same applies to any other physiological aspect of the body, like muscle or the circulatory system, which can be enhanced or stunted depending always on use or neglect; and not only in a singular manner in youth, albeit this is a crucial and defining developmental phase, but also throughout the life of the individual. Like a muscle, language too must be exercised to stay active – the case of a second language illustrates this clearly; as does the age-related condition of cognitive loss, which can be slowed by active use of language and especially multiple languages.
The precedent to the floating realms of sound and thought is the physical intermediation allowed by the senses and the memory matrix, where the neurons of the brain are a receptor and storage for the products of the senses. The sensory mediation is the first degree removed from empirical reality, while the reception and storage of sensory mediation – in the form which here is called “thought” or the “realm of thought” for convenience – is another degree removed – mental intermediation – followed by yet another degree in the synthesis and then recollection of thoughts that the neural matrix makes possible – memory. A further distinction here can be made between passive memory (the unconscious, as Freud calls it) and the activation of accessible memory that is conscious thought. In the case of language, the memory activation is verbalised (rather than via any other motor pairing with thought), but the verbalisation in actual speech is no longer a necessary feature once language is sufficiently established (both in ontogenesis and phylogenesis). This highlights the fundamental and essential constitutive role of language in the initial slicing of the nebulous realm of thought into thoughts in language, although it is incontestable that thoughts and memories can also be formed by the direct input of sensory data of experience – tastes, smells, sounds, etc, which in turn can be reactivated in a more direct sensory path (such as Proust’s madeleine), and which presumably become intermeshed in the storage of thoughts. All of this shows that even in humans the realm of thought really is still a floating realm, especially in the unconscious. This is graphically demonstrated by emotions: an up-welling of floating thoughts but only roughly sliced, which – even perhaps only initially and temporarily – are only roughly paired with sounds and escape or overflow language to find a pathway of expression in other, still floating but less differentiated sounds – like weeping, shouting, etc. – in combination with further physical expression, like motor operations (running or “flight”, striking or “fight”, hormonal responses, orgasm).
On one side in the precedent to the floating realms stands the imputed world of empirical reality, which itself we can only talk about in a way fundamentally delimited by the various degrees of removal – sensory mediation, mental intermediation and memory – culminating in thought and language. To talk about empirical reality in any sort of coherent manner involves the convergence and integration of many different perspectives in one person as well as the perspectives of many people converging and being integrated by a process of experimentation, repetition and mutual consensus (reaching an apotheosis in what is called the scientific convention), all of which are subject to irreconcilable variations, or error, as this irreconcilability of perspective is called in the scientific context. On the opposite side stands the composite matrix of human sensory and neural functions that hosts this consensus of thought, together working in degrees of remove and with overlap to produce the intermediated realm of thought, to which sound is then added by evolution in the double action of slicing sound and thought and pairing them in units. All of this intermediated world of language taken together can be given various names: psychological reality, human reality, constructed reality, intermediated reality, or even simply the reality of language. These are synonymous terms, but none of them are equal to empirical reality; they all require an adjective preceding the word reality, which itself is prefixed for clarity of thought by the adjective empirical. The entire package of human reality is the product of arbitrariness regulated by convention built on differentiation and repetition. In no possible manner can there be any conceivable basis for a correlation between human reality and empirical reality expressed in the formula one to one (1:1), i.e. in the sense of equivalence or reference, as it was termed above in the discussion of referent. The two worlds stand apart, separated by a gulf between them that may appear to be as minuscule physically as a few cells forming the organism’s epidermal membrane, but in effect this gulf is as great as the vast chasm of space between our sun and even the nearest stars, so that the analogy of a human looking through a telescope at distant galaxies comes only vaguely to express the gulf between empirical reality and human reality and the intermediation between them, effected in evolution by the senses, even as these human senses are now perhaps augmented to some degree by technology. Further clouding the picture and calling attention again to the ubiquity of arbitrariness in all things, the very modest perspectives on empirical reality we can muster suggest that empirical reality too is the product of the confluence of any number of arbitrary physical forces and materials in random combinations, so that no element on either side of the gulf is free of the principle of arbitrariness. Given all this, it must surely then become clear to even a child how the signs by which we talk about and understand any form of reality – empirical and human – must themselves be subject to the universal principle of arbitrariness, so that the arbitrariness of the sign is no figure of speech.
The contact zone of the senses between empirical reality and the intermediated reality of the senses is the shaky (i.e. arbitrary in all aspects) foundation on which in turn is built the floating realms of sound and thought. These two realms are the material basis for language and its signs, where a slice of sound is paired with a slice of thought to produce the linguistic sign. Not only is the bond between the two slices arbitrary, but the slicing up of the sound matrix and the thought matrix is also entirely arbitrary. Language allows for a refinement of sound and thought by the resolution of a potential range of sound production and hearing into discreet units (signifiers); and a vast mass of thoughts into smaller and more manageable units (signifieds). In both cases, these units are units of convention – entirely arbitrary. The role of convention has already been discussed in the context of the signifier-signified bond, but convention in all its arbitrariness is just as important in the process of slicing sound and thought that must precede the formation of a signifier-signified bond. Convention plays its part in the slicing process by two actions, namely differentiation and repetition. These are the two underpinning and facilitating forces of convention. They now come to the fore in the discussion. As the foundation stones of convention, differentiation and repetition are the underpinning actions of slicing the realms and combining them in signs. Both differentiation and repetition act via the principle of arbitrariness. The arbitrariness of differentiation is easier to see, but repetition in its own manner relies on arbitrariness inasmuch as repetition is the reinforcement of differentiation.
Differences in units of sound and thought are the foundation of a sign system. As I have previously written (Author 2021, p. 8), the monotone signifier is technically possible, but it would signify only one thing, and that would ultimately be itself. Even a warning siren – which in principle could be a monotone, though rarely is – takes on its meaning in relation to its absence; the turning on of the siren interrupts silence, against which it signifies the onset of danger. In a similar vein, it is technically possible for a sign system to reduce a multiplicity of signifiers to a single signified – after all, this is the character of fascism and totalitarianism – but again it would be a fundamentally uninteresting sign system that signifies in the same way as a monotone sound, despite the multiplicity of signifiers.
A good analogy or comparison is to music – although music (without words) is arguably exactly a system of pure self-signification in which differentiation of sound is the precise point of the system. Perhaps this has its roots in the human heartbeat, itself not only a differentiation between beat and rest, but also between two different phases of the beat (atrial and ventricular – the lub-dub of medical textbooks). By analogy, there is a signified to the signifier of the heartbeat and of music: life itself. The sound of life, to put it most simply, in an intimate and almost – but not quite – direct, unmediated form.
In the system of conventional linguistic signs, however, the set of signifiers is always a series of different sounds: the sounds have to be different from each other in order to become individual slices of sound. The nature of difference in vocal sounds may appear limited in its arbitrariness by the range of the human voice or vocal apparatus, but it must always be remembered that the somatic apparatus co-evolved with language, and so the range and character of the human voice in language produced by the vocal apparatus is the consequence of arbitrariness from the very start. Like all evolutionary traits, the evolution of language as a system of discreet sounds (langue), that are activated and repeated in their multifarious deployment as paroles, is determined by the evolution of the necessary vocal apparatus. Like all evolutionary traits, the vocal apparatus itself proceeds on the basis of selection of specific variations (differentiations) that arise arbitrarily but which accord an advantage to the species, and so are retained in the evolutionary heritage. The capacity for language (in the modern human sense) was not always present in the hominims, and it arose in its modern complexity during a long evolutionary process of interaction between sound and body (especially the brain), so that variation (difference) in sound and also its perception in hearing – as much as the modern catalogue of possible human vocal products – co-determined the evolution of the capacity for language in humans.
By its differentiation, each significant variation (or slice) of sound becomes a potential signifier through the process of repetition. It is not enough for there to be differences in sound (that are in their character entirely arbitrary), but these differences must become relatively fixed in order to become units of speech. They are not immutably fixed, only relatively, i.e. in relation to each other (synchronically) and relative to themselves (in the diachronic view). The sound of a word may change over time (diachronically), and each word allows for a certain degree of variation in itself (pronunciation, accent, dialect, regionalism) before it becomes another word. The language system as a whole also tolerates a certain degree of overlap (such as homophones, homonyms), albeit this type of linguistic efficiency has diachronic origins and relevance, and produces its own kind of difficulty for learning a language, especially learning the written form. Such sound coincidences also offer further proof of the arbitrariness of sound in serving as a signifier. The most extreme example is perhaps the case of the same (written) word existing in different languages, such as angel, which is found in English (an angel with a harp) and Afrikaans and Dutch (the sting of an insect) and also in German (die Angel – a fishing rod or hinge), although their similar pronunciation in the Germanic family differs significantly to the pronunciation of the word in English, where it is of Latin and Greek origin, cognate with engel in Afrikaans, Dutch and German (der Engel). Although direct borrowing in the globalised world of language is now frequently the source of such words today, in the diachronic perspective there will most often be a common root or ancestry – as in evolution itself, where common ancestry leads to common albeit varied physiological traits; here perhaps the cell, the building block of life itself, is the most interesting example of differentiation and repetition.
The slicing of thoughts into signifieds is no less arbitrary than the slicing of sound, and this process relies equally on differentiation, though the difference is much harder to identify and recognise than difference in sound. It is for this reason that I am inclined to view sound – and the specific system of vocal sound that is language – as being the necessary constitutive condition for the development of the realm of thought in the evolutionary perspective. This becomes clearer with each additional degree of abstraction in language and thought, so that the early correlation (in evolution as much as in a child’s development) of a sound (e.g. a cry of pain or joy) with a thought (pain or joy felt) may appear less arbitrary than the linguistic and mental complexity of a discourse on language in language, for example, where each word or concept is in need of clarification and definition to establish some sort of common agreement (convention) about the slices of thought in question. Even with language, a common understanding is only the result of a shared convention, and the success of understanding is never a given, even with the best efforts of language users to engage the convention faithfully. Without language, the floating realm of thought is doomed to remain cloudy and impenetrable. This is not to deny that the realm of thought can exist without language. Quite the contrary. Animals without a language system comparable to human language still clearly have brains – the matrix for thinking – and they show significant evidence of the ability to learn, remember and activate thoughts as motor responses, even associated with sound production, within the range of their species-specific vocal apparatus. Another major example: humans have retained the floating realm of thought in a form that exists with – at best – a weak bond to the defining slices of language. This in the unconscious, which is arguably still the most powerful and important element of the human mental apparatus.
Language, and the differentiation of thought that language makes possible, can be said to coexist with the non-linguistic or pre-linguistic cloud of thought that is the unconscious and out of which language arises. More than that, language itself is structured in this way by the unconscious (Author 2025, pp. 133-137).
Conclusion
In overview, the arbitrariness of the sign can be seen in all aspects of the sign: in the arbitrary bond internal to the sign between the signifier and the signified; in the absence of a referent; in the floating realms of sound and thought out of which the elements of the sign are carved; in the evolutionary precedents to the floating realms; in the differentiation of a sign from all other signs at the levels of sound and thought; and in the social adoption and repetition of the sign that is convention. Certain signs may be derived from other signs, but this derivation (sometimes called motivation) does not detract from the original arbitrariness. There can be various types of affinity between signs, at the level of sound or thought, or of the sign in its totality with other signs.
A dominant and persistent mythology privileges the communication theory of language (in particular) and signs (in general). Although always imprecise, provisional and approximate, the communication function of language is not in dispute, and it certainly underpins the evolutionary foundation of language and sign systems. However, communication is only one of several functions that language performs, albeit an early function in both ontogenetic and phylogenetic time; and communication as a function is certainly not limited to the domain of language – spoken or written – or even sign systems. Direct expression and response on the body remain important in human communication, even as direct human contact is being increasingly renounced in the 21st century and replaced by the intermediation of machines, such as social media and similar surrogates and – more terrifyingly – new autonomous and remote machines of war. While the mnemonic function of language is increasingly neglected today, it was at least as important as communication in the emergence of human language. The mnemonic endowment of language was a radical step forward in human evolution that allowed for the collection, storage and transfer of knowledge not only directly between individuals (i.e. communication proper), but in the function of bridging different physical locations and the temporal gap between generations, and, perhaps most importantly, in the stimulation and growth of the internal neural cognitive network, which ultimately produced human consciousness. The mnemonic function of language breaks the tyrannical hold that experience previously maintained over knowledge in the early stages of human evolution.
The importance of memory – the capacity to remember which language facilitates – in the physical evolution of the brain to become the outsized neural matrix that distinguishes modern humans cannot be emphasised enough, and this function of language becomes pre-eminent in the drive towards the development of writing systems. Given the power of language to constitute human reality, it will be interesting to see the compounding effect of so-called artificial intelligence (AI) in the atrophy of language and of the brain, a development already set in motion by large-scale digital intermediation. The metaphorical Tower of Babel will come to seem like a paradise compared to the future monolingual dominance of the world by a handful of simplified languages endlessly cannibalising their written cultures and spewing out vacuous reams of degraded hallucinatory verbiage. With these technologies, the human disavowal of empirical reality has the potential to exceed all previous levels of excess and reach an irreversible state of psychosis.
The functions of language in the evolutionary perspective are subject to the human bifurcating tendency to exceed the evolutionary function of a trait and to endow it with additional or surplus functions that serve no direct or obvious supportive evolutionary purpose but which nevertheless persist in the species – if for no other reason than that they are not an obstruction to survival, at least so far. Speculatively, it may be possible to suggest that they could even have reinforced the evolutionary purpose previously, to produce unique features in homo sapiens, where the persistence of surplus functions may be facilitated via the path of nature or via socialisation, or by a combination of both. This is more than self-evident in the traits of sexuality and labour, and pre-eminent in the case of language, the crowning social dimension of humanity.
To approach a discussion of language in terms of only a single function – as is the style of linguistics and the philosophy of language – is to atomise and compartmentalise functions, and then to nominate one overly-simplified function as the only or primary function. This is an error of analysis; not only does it reduce language to just one of its part components but it wilfully ignores the importance of language as a whole in human evolution, and – more significantly – the arbitrary character of language and its signs as the defining and irrevocable constitutive element of human language. The chief consequence of recognising the arbitrariness of the sign is a heightened sensitivity to the necessity for acknowledging and interrogating perspective when interpreting the sign. To recognise and understand the arbitrariness of the sign means also to recognise and understand the arbitrariness of perspective and nomination, and the arbitrariness of convention itself. Underlying all these human components of the sign is the utter arbitrariness of evolution, and of empirical reality. Disavowal of arbitrariness in any of these dimensions is a slippery anthropomorphic slope to the denial of empirical reality.
References
Abramova, E. and Fernandez, R. 2016. Questioning arbitrariness in language: A data-driven study of conventional iconicity. Proceedings of the 2016 conference of the North American chapter of the association for computational linguistics. Stroudsburg: Association for Computational Linguistics.
Barthes, R. 1967. The death of the author, trans. by R. Howard and first published in the English version (not French) in Aspen Magazine: The Minimalism Issue (#5,6), New York City, USA: Roaring Fork Press.
Benveniste, E. 1939. Nature du signe linguistique. Acta Linguistica. 1(1): 23-29.
Bloomfield, L. 1933. Language. New York: H. Holt and Company.
Bolinger, D. 1949. The Sign Is Not Arbitrary. Boletín del Instituto Caro y Cuervo 5: 52-62. Reprinted in Bolinger, Forms of English: Accent, Morpheme, Order. ed. by I. Abe and T. Kanekiyo. Cambridge: Harvard University Press and Tokyo: Hokuou, 1965.
Cuypere, L. de. 2008. Limiting the iconic: From the metatheoretical foundations to the creative possibilities of iconicity in language. Amsterdam: John Benjamins Publishing Company.
Danesi, M. 2022. Saussure’s View of Language as a System of Arbitrary Signs. Language and Semiotic Studies. 8(1): 16-23.
Dingemanse, M., Blasi, D.E., Lupyan, G., Christiansen, M.H. and Monaghan, P. 2015. Arbitrariness, iconicity, and systematicity in language. Trends in Cognitive Sciences. 19(10): 603–615.
Downing, L. and Stiebels, B. 2012. Iconicity. In J. Trommer The morphology and phonology of exponence. Oxford: Oxford University Press.
Foucault, M. 1969. Qu’est-ce qu’un auteur? being the text of a seminar paper presented on 22 February at the Société Française de Philosophie, published version in Société Française de Philosophie Bulletin 63 (3): 73-104. A slightly modified version presented in 1970 at Buffalo University (USA), trans. by J. Harari and published as What is an author? in Textual Strategies: Perspectives in Post-Structuralist Criticism. Harari, J.V., ed. (Ithaca: Cornell University Press, 1979).
Freud, S. 1905. Drei Abhandlungen zur Sexualtheorie (Franz Deuticke: Liepzieg and Vienna), trans. by J. Strachey as Three essays on the theory of sexuality in The standard edition of the complete psychological works of Sigmund Freud 7: 125-245, ed. by J. Strachey (London: Hogarth Press, 1953-1974).
_______. 1915. Das Unbewusste (Int. Z. Psychoanal. 3(4): 189-203 and (5): 257-69), trans. by C.M. Baines and J. Strachey as The unconscious in Standard ed., 14: 159–204.
_______. 1927. Fetischismus (Almanach der Psychoanalyse. 1928: 17-24), trans. by J. Riviere and J. Strachey as Fetishism in Standard ed., 21: 148-157.
_______. 1933. Neue Folge der Vorlesungen zur Einführung in die Psychoanalyse (Vienna: Internationaler Psychoanalytischer Verlag), trans. by J. Strachey as New introductory lectures on psycho-analysis. Standard ed., 22: 1–182.
Gamkrelidze, T.V. 1974. The Problem of “L’arbitraire du signe”. Language. 50(1): 102–110
Gvoždiak, V. 2022. Arbitrariness and Rationality. Language and Semiotic Studies. 8(1): 45-54.
Holdcroft, D. 1991. Saussure: Signs, Systems, and Arbitrariness. Cambridge: Cambridge University Press.
Honders, E and Honders, E. 1982. A Survey of Iconic Aspects in Language. Interlanguage Studies Bulletin. 6(1): 3-21.
Jakobson, R. 1957. Shifters, verbal categories, and the Russian verb. Cambridge: Russian Language Project, Dept. of Slavic Languages and Literatures, Harvard University.
_______. 1960. Linguistics and poetics. in T.A. Sebeok. Style in Language. New York: The Technology Press of Massachusetts Institute of Technology and John Wiley and Sons.
_______. 1965. Quest for the Essence of Language. Diogenes. 13(51): 21–37.
_______. 1985. Verbal Art, Verbal Sign, Verbal Time. Minneapolis: University of Minnesota Press.
Kant, I. 1790. Kritik der Urteilskraft (Berlin: Lagarde und Friedrich), trans. by J.C. Meredith as Critique of Judgement (Oxford: Oxford University Press, 2007).
Lakoff, G and Johnson, M. 1980. Metaphors We Live By. Chicago: University of Chicago Press.
Marx, K. 1847. Misère de la philosophie. Réponse à la Philosophie de la misère de M. Proudhon. Paris: A. Frank, and Brussels: C.G. Vogler.
_______. 1867. Das Kapital – Band 1 Kritik der politischen Ökonomie (Hamburg: Verlag von Otto Meisner), trans. by S. Moore and E. Aveling as Capital – Volume 1 in Volume 35 of Marx/Engels Collected Works (New York: International Publishers, 1996).
_______. 1863-1883. Das Kapital – Band 3 Kritik der politischen Ökonomie (first published Hamburg: Verlag von Otto Meisner, 1894), trans. by E. Untermann as Capital – Volume 3 in Volume 37 of Marx/Engels Collected Works (New York: International Publishers, 1998).
Nanny, M. and Fischer, O. 1999. Form Miming Meaning. Philadelphia: John Benjamins.
Nietzsche, F. 1885-87. Sämtliche Werke Kritische Studienausgabe Band 12: Nachgelassene Fragmente 1885-1887, ed. by Giorgio Colli and Mazzino Montinari (Berlin: Walter de Gruyter & Co., 1967-77).
Author. 2021.
_______. 2025.
Peirce, C.S. i. 1893 – 1902. The icon, index and symbol. Reprinted in Collected papers of Charles Sanders Peirce. ed. by C. Hartshorne & P. Weiss, vol. 2: Elements of logic. Cambridge: Harvard University Press.
_______. i. 1897 – 1910. Logic as semiotic: the theory of signs. Reprinted in Philosophical Writings of Peirce. ed. by J. Buchler. New York: Dover, 1955.
Perniss, P., Thompson, R.L., and Vigliocco, G. 2010. Iconicity as a general property of language: Evidence from spoken and signed languages. Frontiers in Psychology. 1: 227.
Plato. c. 384-3 BC. Cratylus trans. by C.D.C. Reeve. Indianapolis and Cambridge: Hackett, 1997.
_______. c.385-370 BC. The Symposium trans. by M.C. Howatson. Cambridge: Cambridge University Press, 2008.
Saussure, F. de. i.1906-1911. Lectures delivered at the University of Geneva and published from auditors’ notes by C. Bally and A. Sechehaye under the title Cours de linguistique générale (Paris: Payot, 1916), trans. into English by W. Baskin as Course in General Linguistics (New York: The Philosophical Society, 1959) and subsequently edited by P. Meisel and H. Saussy (NY: Columbia University Press, 2011); references are to this edition, abbreviated as Course. Related unpublished manuscripts by Saussure discovered posthumously are published as Écrits de linguistique générale (Paris: Gallimard, 2002, edited by S. Bouquet, R. Engler and A. Weil) and trans. into English by C. Sanders and M. Pires as Writings in General Linguistics (Oxford: OUP, 2006); references are to this edition, abbreviated as Writings.
Sebeok, T.A. 1994. Signs: An introduction to semiotics. Toronto: University of Toronto Press.
Shklovsky, V. 1917. “Iskusstvo kak priyom” in Sborniki po teorii poeticheskovo yazyka III [Studies in the Theory of Poetic Language III] (Petrograd: Tipografiya Sokolinskogo, 1919), trans. by L.T. Lemon and M.J. Reis as “Art as Technique” in Russian Formalist Criticism: Four Essays, ed. L.T. Lemon and M.J. Reis (Lincoln: University of Nebraska Press, 1965).
Silverstein, M. 1976. Shifters, linguistic categories, and cultural description. In K. Basso & H. Selby, eds., Meaning in Anthropology. Albuquerque: University of New Mexico Press.
Simone, R. 1995. Iconicity in Language. Philadelphia: John Benjamins.
Stawarska, B. 2015. Saussure’s Philosophy of Language as Phenomenology. Undoing the Doctrine of the “Course in General Linguistics”. Oxford: Oxford University Press.
_______. 2020. Saussure’s Linguistics, Structuralism, and Phenomenology. Cham: Springer Verlag.
Stoltz, D.S. 2021. Becoming a dominant misinterpreted source: The case of Ferdinand de Saussure in cultural sociology. Journal of Classical Sociology. 21(1): 92-113.
Thompson, R.L., Vinson, D.P. and Vigliocco, G. 2012. The road to language learning is iconic: Evidence from British Sign Language. Psychological Science. 23(12): 1443–1448.
Thomas Olver is an independent researcher and translator living in Pretoria, South Africa. He studied modern languages, comparative literature and psychology (Pretoria, Witwatersrand, Zurich). He favours a collaborative approach to epistemology. His research and teaching interests include semiotics, psychoanalysis, aesthetics, narratology, modern literature, translation and the history of ideas. His most recent publication is a monograph on Freud’s theory of bisexuality (Radical Freud, Routledge 2025; www.routledge.com/9781032812564). Full list of publications available on www.researchgate.net/profile/Thomas-Olver-2